# Predicting the Type and Target of Offensive Posts in Social Media

Source: [Predicting the Type and Target of Offensive Posts in Social Media](<https://www.sota2.com/research/paper/68fa659f1e209131cea0b49f>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:30:43.473Z

Authors: Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Sara Rosenthal, Noura Farra, Ritesh Kumar

Published: 2019-02-25T23:54:40.000Z

## Abstract

As offensive content has become pervasive in social media, there has been much research in identifying potentially offensive messages. However, previous work on this topic did not consider the problem as a whole, but rather focused on detecting very specific types of offensive content, e.g., hate speech, cyberbulling, or cyber-aggression. In contrast, here we target several different kinds of offensive content. In particular, we model the task hierarchically, identifying the type and the target of offensive messages in social media. For this purpose, we complied the Offensive Language Identification Dataset (OLID), a new dataset with tweets annotated for offensive content using a fine-grained three-layer annotation scheme, which we make publicly available. We discuss the main similarities and differences between OLID and pre-existing datasets for hate speech identification, aggression detection, and similar tasks. We further experiment with and we compare the performance of different machine learning models on OLID.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Offensive Language Identification on OffensEval Danish (test)](<https://www.sota2.com/research/sota/offensive-language-identification-on-offenseval-danish-test.md>): Offensive Language Identification; OffensEval Danish (test); Macro F1: 0.699.

- [Offensive language identification on OLID English Sub-task A 1.0 (test)](<https://www.sota2.com/research/sota/offensive-language-identification-on-olid-english-sub-task-a-1-0-test.md>): Offensive Language Identification; OLID English Sub-task A 1.0 (test); Macro F1: 0.728.

- [Categorization of offensive language type on OLID English Sub-task B](<https://www.sota2.com/research/sota/categorization-of-offensive-language-type-on-olid-english-sub-task-b.md>): Categorization of offensive language type; OLID English Sub-task B; Macro F1: 0.593.

- [Offensive language target identification on OLID English 2019 (test)](<https://www.sota2.com/research/sota/offensive-language-target-identification-on-olid-english-2019-test.md>): Offensive language target identification; OLID English 2019 (test); Macro F1: 45.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1902.09666>)

[PDF](<https://arxiv.org/pdf/1902.09666.pdf>)
